Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

Buckling analysis of cantilever pultruded

I-sections using
Kunj J. Patel1, Satyen D. Ramani2,
1

M.E. student, Civil Engineering Department, SAL institute of technology and engineering research,
Gujarat, India.
2
Asst. Professor, SAL institute of technology and engineering research, Gujarat, India.

ABSTRACT
For steel beam buckling analysis is a critical area of study to determine overall section capacity subjected to
bending. Pultruded I-beams are manufactured using fiber and matrix composite lamina hence is usually orthotropic
compared to conventional steel I-beams. Many researchers have studied the buckling characteristics of simply
supported pultruded FRP beams for various types of loadings. In this paper an attempt has been made to study the
buckling characteristics of cantilever beam subjected to point load at free end for WF(wide flange) & NF(narrow
flange) section. The validation has been made with experimental data. Further the parametric study has been
carried out to study the effect of fiber orientation along with fiber volume fraction on critical buckling loads.

Keyword: - Pultruded I-beams, Fiber orientation, buckling characteristics, fiber volume fraction.
1. INTRODUCTION:By decades the use of composite material is increased. By good knowledge of the fundamental properties of
composites the rise of composites can be explained.FRP profiles have been used over more over past few years, due
to high ratio of strength to weight. In the construction industry, recent applications have shown the structural and
cost efficiency of FRP structural shapes, such as thin-walled open sections made through so-called pultrusion
process. FRP beam and column offers great benefits in placement of composites with the help of lighter machine
and initial haulage. It also offers the less schedule of maintenance and offer easy attachment of fittings, with FRP
structures being engineered for a long service life. The physical and chemical properties of the composites make
them a very attractive material option for selection particularly in more corrosive environments where traditional
materials are known to deteriorate. To realize these benefits, the world of FRP requires demystifying, and the
selection of these products made more easily understood. Pultruded profiles for commercial and structural
applications which are produced by many manufacture the world. Most manufacturers produce custom profiles for
commercial applications. A number of industry groups represent and loosely coordinate the activities of pultrusion
manufacturers. Leading groups are the Pultrusion Industry Council of the European Pultrusion Technology
Association (EPTA) and the American Composites Manufacturers Association (ACMA). Commonly produced
pultruded profiles are usually available as wide-flange, I, rectangular tube, channel, square tube, plate materials and
angle profiles. Standard profiles range from width to approximately 12 in. (300 mm) and 2 in. (50 mm) in height and
have material thicknesses of two in. (6 to 13 mm).At this time there is no proper guidelines for the design of framed
structures using either pultruded structural profiles as there are for concrete structures strengthened with FRP
strengthening systems or reinforced with FRP rebar. However, two design manuals are developed: The Euro comp
Design Code and Handbook (Euro comp, 1996) and The Structural Plastics Design Manual (ASCE, 1984).

2. LITERATURE STUDY:M.M.Alinia, A.Dibaie, et al [1] studied buckling behavior of I-beams. The typical plated columns, one is 1m height
and second is 4m height having similar section properties are considered. The web and flange width are assumed to
be 1000mm and 300mm. The web and flange thickness is carried out 5mm and 30mm. During parametric studies for
web and flange slenderness ratio, thickness is varied. Selections of flange and web thickness are as per limits
defined in AISC360.The ABAQUS software is used in incremental nonlinear large displacement push over analysis.

1146

www.ijariie.com

71

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

The nonlinear push over analysis is carried out by utilizing the modified Risk method, allowing for both softening
and stiffness hardening of plated structures.The structural elements are taken as four nodes doubly curved shell
element S4R in ABAQUS. This element accounts for large rotation and finite membrane strains. The S4R element
has three rotational and three translation of freedom at each node.it is based on isoperimetric formulation and uses
one integration point on its mid-surface to form the element internal force vector. The reduced integrated
formulation is used to reduce time and provide accurate results.
Ever J. barbero & Ioannis G. Raftoyiannis et al [2] has studied pultruded I-beams subjected to various loading
conditions and their elastic buckling modes. The lateral and distortional buckling coupling to appear in thin-walled
cross verify to their predict bending-twisting coupling and shear effects are accounted. The effect of volume fraction
and fiber orientation in the matrix is taken into account for parametric study. The buckling strength determine as
overall strength of the member. The energy criterion for equilibrium of structure is used for theory approach. The
solution of the problem is attempted by Rayleigh-Ritz method. The pre buckling solution is derived using laminated
beam theory. Coupling of local and lateral buckling modes always occurs due to the low stiffness of the material in
the perpendicular direction to the beam axis. For high depth-width ratio, lateral buckling is the governing mode of
failure. For low height-depth ratio coupled local and distortional buckling results in a reduced section. The simply
supported beams with free warping in both the ends and having concentrated load at the centroid is considered in the
specific cases, the theoretical formulation and the solution methodology can be easily applied for any boundary
conditions.
Rami Haj-Ali, HakanKilic et al [3] has invested Coupon tests and they are used to calibrate 3D micromechanical
models and to verify their prediction for the non-linear elastic behaviour of pultruded FRP composites. The material
system is made from E-glass/vinyl ester composite plate with both continuous filament mat (CFM) layers and glass
roving layers. Tension, shear and compression tests were performed, using off-axis coupons cut with different
roving reinforcement orientation. The overall linear elastic properties are identified along with the nonlinear stressstrain behaviour under in-plane multi-axial tension and compression loading. The tests were carried out of coupons
with off-axis angles: 0, 15, 30, 45, 60 and 90, where each test was carried out three to five times. Finite element
analysis is used to examine geometry condition end-clamping condition and axis orientation. Lower initial elastic
modulus and softer nonlinear stress-strain responses observed in the tension compared to those in compression for
all axis orientation. The nonlinear behaviour can be attributed to the relatively low overall fiber volume fraction in
pultruded composites and manufacturing defects such as micro cracks and voids. End coupling effect for the tested
geometry is small at the centre and allows extracting the non-linear stress-strain response of anisotropic material.
Shear tests are used to calibrate the in situ nonlinear elastic properties of the matrix. The material has a lower initial
stiffness and lower ultimate tensile stress in tension compared to the corresponding values of compression tests,
regardless of thickness and angles approaching angle of 90. Good scalability of material and geometry appear to
exist between the and in. thickness coupons tension and compression of axis (45) tests can be used to
determine the initial shear modulus of FRP pultruded material.
Lawrence C. Bank, et al [4] has considered the shear stiffness of pultruded FRP beam.. A new material stiffness
property called shear modulus is taken into account for the shear deformation effects. Results of an experimental
program conducted to obtain the section shear modulus for variety of FRP beams. The analysis of a rigid portal
frame structure is presented to demonstrate the influence of the beam section stiffness properties on frame deflection
and forces. The stiffness property can be obtained for standard putruded beams using test methodology that gives
beam flexure modulus. Using the stiffness properties designers can take account for the effect of shear deformation.
J. Lee, S.-E. Kim, K. Hong et al [5] has studied lateral buckling of a laminated composite I-section beam. General
analytical model applicable to the lateral buckling of section subjected to various loading types is developed. The
model is based on classical lamination theory. It accounts for the material coupling for arbitrary laminate stacking
sequence and various boundary conditions. The effect of the location of applied load on buckling capacity is also
included. A finite element model and a displacement model is developed to predict buckling modes and critical
loads of a thin walled composite beam with boundary conditions. Numerical results are obtained under, uniformly
distributed load; pure bending and central point load with angle-ply laminates. The effects of location of applied
load, types of loads and orientation of fiber are studied. Lateral buckling capacity of beams with transverse loads is
affected by the fiber orientation as well as location of loads. The maximum buckling loads occurs near 45 fiber
angle for all cases of span to height ratio. For beam under pure bending with off-axis fiber orientation, orthotropic
closed-form solution is not appropriate for lateral buckling loads due to existence of coupling stiffness.

1146

www.ijariie.com

72

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

Pizhong Qiao, Julio F. Davalos, Jialai Wang et al [6] has carried out an analytical study of local buckling of
discrete laminated panels or plates FRP structural shapes. Flanges of FRP shapes are modelled as discrete panels
subjected to uniform axial loads. Two cases with different boundary conditions and elastic restraints of composite
plates are studied. Using the regression analysis simplified expressions for predictions of plate buckling stress
resultants are formulated in terms of coefficients of boundary elastic restraints. The effects of restraint at the flange
web connection are considered. The theoretical predictions show good agreement with experimental data and FEM
eigen value analysis for local buckling of FRP columns. The present formulation can be applied to several cases to
determine buckling capacities.

Fig-1: Pultruded profiles: I, wide flange, angle, tube, and channel

3. VALIDATION STUDY:3.1 Material Geometry & properties:In this study, two wide flange profiles WF 0.1 x 0.1 x 0.006m (4x4x1/4), WF 0.15 x 0.15 x 0.0095m (6 x6 x3/8),
and two narrow flange profiles NF 0.1 X 0.2 X 0.0095m (4x8x3/8), NF 0.07 X0.15 X 0.0095m (3x6x3/8) of FRP
I-beams, which were manufactured by the pultrusion process and provided by Creative Pultrusions, Inc., Alum
Bank, PA, are taken. The 60% E-glass fibers and 40% polyester resin is used in pultrusion process.The results of
critical buckling loads for all profiles are compared to the experimental data from research work carried out by
Luyang shan et al [7]. The lamina properties of section are given as follows.
Table -1: Material Lamina properties
Exx

1.14 x 1010 Pa

Eyy

4.47 x 1010 Pa

Ezz

1.14 x 1010 Pa

NUxy

0.272

NUyz

0.272

NUxz

0.046

Gxy

4.20 X 109 Pa

Gyz

4.20 X 109 Pa

Gxz

4.20 X 109 Pa

3.2 Modelling in ANSYS:Each profile is modelled as cantilever beam with end condition (Ux=Uy=Uz=0) at all nodes of the flange and web at
one end to simulate the experimental setup[7].Vertical node at mid height of web is applied and eigen buckling
analysis is carried out.Fy=10 is applied as an initialization load for eigen buckling. Modelling is done using shell

1146

www.ijariie.com

73

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

181, special element which facilitates the laminate analysis with orthotropic and anisotropic lamina stacking
arrangement material. For the analysis the beam length is taken 3m.Global lateral torsional buckling is observed as a
critical buckling mode.corrosponding loads for 1 st mode of lateral torsional buckling for each of four section
analysed is than compared with experimental data value obtained by Luyang Shan et al[7].
3.3 Critical buckling load (Pcr):-

10000
9000
8000
7000
6000
Pcr(N)

5000

luyang shan

4000

ANSYS

3000
2000
1000
0
I 0.10X0.10

WF 0.15X0.15

I 0.07X0.15

I 0.10X0.20

Fig-2: Critical buckling load comparison

Table -2: Critical buckling load comparison between ANSYS and study done by Luyang Shan

WF 0.1m X 0.1m X0.006m

WF0.15mX0.15mX0.0095m
I 0.07mX0.15mX 0.0095m
I 0.10m X0.2mX 0.0095 m

Luyang shan

Pcr (N)
1436

ANSYS

1435

Luyang shan

8624

ANSYS

8858

Luyang shan

2174

ANSYS

2000

Luyang shan

4503

ANSYS

4200

Table-2 shows the comparision of critical buckling load from ANSYS & Luyang Shan. it can be concluded that the
results are very well in the compliance range with experiment.

1146

www.ijariie.com

74

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

4. FIBER ORIENTATION STUDY:One narrow flange section (NF 0.1 x 0.2 0.0095 m) and one wide flange section (WF 0.15 x 0.15 x 0.0095 m)
having different fiber volume fraction are taken as cantilever beam for study. Vertical 1 N load is applied on the
middle node of web at free end. Material properties are taken as described in Table-1. The effect of different fiber
orientation for different fiber volume fraction for cantilever beam is studied. Further in the study, all the possible
buckling modes are examined to find the critical load values corresponds to global lateral torsional buckling mode
(ltb) & local flange buckling mode (lfb). Critical load corresponds to lateral torsional buckling is denoted as P0 &
local flange buckling load is denoted as PL. Buckled shape for a typical cantilever I-beam subjected to point load at
free tip is shown for both case ltb and lfb in Fig-3.

Fig-3: Global and local buckling modes

Table -3: Critical buckling loads for different fiber angles for cantilever beam (NF-profile)
Fiber volume Fraction = 15%

1146

P0/PL

Fiber Angle

P0(N)

PL(N)

15

1614

12706

0.127027

30

1526.6

14330

0.106532

45

1404.4

15650

0.089738

60

1286.7

14286

0.090067

75

1190.1

12581

0.094595

www.ijariie.com

75

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

Fiber Volume Fraction = 30 %

P0/PL

Fiber Angle

P0(N)

PL(N)

15

2570.4

17379

0.147903

30

2373.9

19428

0.12219

45

2146

20923

0.102567

60

1923.8

18687

0.102949

75

1743.8

16911

0.103116

P0(N)
4899.9
4559.1
4129.5
3702.8
3354.2

PL(N)
33881
38212
41425
37068
33097

P0/PL

Fiber Volume Fraction = 60%

Fiber Angle
15
30
45
60
75

0.144621
0.119311
0.099686
0.099892
0.101345

Where, P0 = Critical buckling load for 1 st mode of global lateral torsional

buckling.
PL = Critical buckling load for 1st mode of Pure Local flange buckling.

Vf 15
18000
16000
14000
12000
P(N)

10000
8000

P0(N)

6000

PL(N)

4000
2000
0
0

20

40

60

80

Fiber Angle (degree)

1146

www.ijariie.com

76

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

Vf 30
25000
20000
15000
P(N)
P0(N)

10000

PL(N)
5000
0
0

10

20

30

40

50

60

70

80

Fiber Angle (degree)

Vf 60
45000
40000
35000
30000
25000
P(N)
20000
15000
10000
5000
0

P0(N)
PL(N)

20

40

60

80

Fiber angle(degree)
Fig-4: fiber Angle vs Critical buckling load (Cantilever beam-NF profile)

0.16
0.14
0.12
0.1
P0/PL 0.08
0.06
0.04
0.02
0

Vf 30
Vf 15
Vf 60

20

40
Fiber Angle (degree)

60

80

Fig-5: Fiber Angle vs P0/PL (Cantilever beam-NF profile)

1146

www.ijariie.com

77

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

Table -4: Critical buckling loads for different fiber angles for cantilever beam (WF-profile)
Fiber Volume Fraction = 15 %
Fiber Angle

P0(N)

PL(N)

P0/PL

15

3421.6

6093.4

0.56153

30

3241.2

6789.1

0.47741

45

2975.2

7325.7

0.40613

60

2720.9

6724.4

0.40463

75

2543.2

6004.5

0.42355

Fiber Angle

P0(N)

PL(N)

P0/PL

15

5424

8473.3

0.64013

30

5070.1

9432.4

0.53752

45

4563.5

10105

0.45161

60

4081.2

9055.2

0.4507

75

3742.3

8210.7

0.45578

Fiber Angle

P0(N)

PL(N)

P0/PL

15

10351

16495

0.62752

30

9727.9

18473

0.5266

45

8775.5

19889

0.44122

60

7849.6

17860

0.43951

75

7191.8

16040

0.44837

Fiber Volume Fraction = 30%

Fiber Volume Fraction = 60%

Where, P0 = Critical buckling load for 1 st mode of global lateral torsional

buckling.
PL = Critical buckling load for 1st mode of Pure Local flange buckling.

vf 15
8000
6000
P(N) 4000
P0(N)

2000

PL(N)
0
0

20

40

60

80

Fiber Angle(degree)

1146

www.ijariie.com

78

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

Vf 30
P0(N)

12000

PL(N)

10000
8000
P(N)

6000
4000
2000
0
0

10

20

30

40

50

60

70

80

Fiber Angle(degree)

Vf 60
25000
20000
15000
P(N)
10000

P0(N)

5000

PL(N)

0
0

20

40

60

80

Fiber Angle(degree)
Fig-6: fiber Angle vs Critical buckling load (Cantilever beam-WF profile)
0.7
0.6
0.5
0.4
P0/PL

Vf 30

0.3

Vf 15

0.2

Vf 60

0.1
0
0

20

40

60

80

Fiber Angle (degree)

Fig-7: Fiber Angle vs P0/PL (Cantilever beam-WF profile)

1146

www.ijariie.com

79

Vol-1 Issue-2 2015

IJARIIE-ISSN (O)-2395-4396

5. RESULTS & CONCLUSION:The increase in PL can be explained due to increase of in plane shear resistence as the fiber angle increses. The
minimum inplane buckling resistance is obtained for fiber orientation (45/0/+45). Fig 5 & 7 shows the variation
P0/PL with respect to fiber orientation which suggests that for (-45/0/+45),the P0/PL is minimum. From the
strength aspect the value of P0/PL should be minimum for optionized utilization of section strength.For fibr
orientation being constant, the increase in fiber volume leads to increase of both P0 and PL which is in due
cmpliance of other literature studies.[3,5].
Hence it can be concluded that for cantilever pultruded I-beam, the fiber orientation can be improve local flange
buckling capacity but will not improve global lateral torsional buckling capacity.

REFERENCES:[1] M.M.Alinia, A.Dibaie, Buckling and failure characteristics of slender web I-column girders under interactive
compression and shear, comp. meth. Civil Eng. Vol.3, 1(2012) 15.
[2] Ever J. barbero & Ioannis G. Raftoyiannis, Lateral and distortional buckling of pultruded I-beams.,
ELSEVIER science limited composite structures 27 (1994) 261-268
[3] Rami Haj-Ali, HakanKilic, Non-linear behaviour of pultruded FRP composites., ELSEVIER science limited,
Composites: Part B 33 (2002)173-191.
[4] Lawrence C. Bank, Properties of Pultruded Fiber Reinforced Plastic Structural members, Transportation
research record 1223.
[5] J. Lee, S.-E. Kim, K. Hong, Lateral buckling of I-section composites beams, Elsevier Science Ltd.,
Engineering Structures 24 (2002) 955-964.
[6] Pizhong Qiao, Julio F. Davalos, Jialai Wang, Local buckling of composite FRP shapes by discrete plate
analysis, journal of structural engineering, vol.127, no. 3, march, 2001.
[7] Luyang Shan, Explicit buckling analysis of Fiber Reinforced Plastic (FRP) composite structures,Washinton
state university,May,2007.

1146

www.ijariie.com

80